I have a form for requests; check it out if you want to request a character.
The female student uniform from Fire Emblem: Three Houses. Trained on NAI. Tested at 0.9-1.0 strength.
Contains two variant versions. Add "gmuniform" to have your character wear either variant, and tags like "blue thighhighs" or "long sleeves" may help.
Type 1 is the variant worn by Lysithea, Marianne, and Annette, among others.
Type 2 is the variant worn by Hilda, Petra, and Leonie.
My 400th upload. Thanks to everyone that's been along for the ride.
Check out my Discord if you want to chat about video games (or yell at me about requests.)
Description
For the Hilda fans.
FAQ
Comments (9)
Crazy timing! For months I'd been hoping for a Lora that does this, and today decided I'd make one myself. While figuring out how to do so, this was uploaded. Looks like mine's been outclassed before I even finished making it.
Thanks, novowels! Out of curiosity: what's the best approach when training a uniform/outfit lora like this?
Simply avoid repetition of the characters wearing it? Cut the head out of frame or erase the facial features? I mean, in order to minimize the character test data bleeding into the uniform.
Just tagging the appearances of the characters in the dataset as well as possible seemed good enough. It did help that this uniform is worn by several different characters, so there was some variety in the dataset.
I need Eirika with Garreg Mach Uniform, please.
Sure, I've added it to the gallery.
Excellent LoRa, my respectsHave you thought about making uniforms for men? It would be a great addition
I have actually thought about this but there is more variation among the male students' uniforms, not to mention that there is just less art of the men in general. I'll look into it eventually.
@novowels Well anyway, then I'll try to do it myself, If you use the character Lora with that uniform you could make your job easier by feeding Stable-diffusion
Goated
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.